{ "background": "Power-distribution systems in many developing nations suffer from significant technical and commercial losses, reducing overall yield. Existing evaluation methods for improvement technologies often rely on modelled or non-comparative field data, lacking rigorous causal inference for real-world performance.", "purpose and objectives": "This article presents a novel methodological framework for conducting a randomised field trial (RFT) to empirically evaluate the impact of specific equipment interventions on power-distribution system yield. The objective is to establish a robust protocol for comparative performance measurement under actual operational conditions.", "methodology": "The methodology designs a cluster-randomised trial where primary substations serve as the unit of randomisation. Matched pairs are formed based on baseline technical loss profiles and customer density. One substation in each pair is randomly assigned to receive the intervention (e.g., advanced conductor or transformer). The core analysis employs a differences-in-differences model: $Y{it} = \\beta0 + \\beta1 \\text{Treat}i + \\beta2 \\text{Post}t + \\beta3 (\\text{Treat}i \\times \\text{Post}t) + \\epsilon{it}$, where $Y_{it}$ is percentage yield. Inference uses cluster-robust standard errors.", "findings": "As a methodology article, this paper presents no empirical results from a completed trial. However, the proposed design and power calculation indicate that, for a hypothesised yield improvement of 3–5 percentage points, a minimum of 30 substation clusters per trial arm provides 80% power to detect an effect at the 5% significance level, accounting for intra-cluster correlation.", "conclusion": "The structured RFT methodology provides a rigorous, replicable framework for evaluating engineering interventions in power distribution, moving beyond anecdotal or purely modelled assessments to generate credible evidence of causal impact.", "recommendations": "Utilities and researchers should adopt randomised field trial designs for capital project evaluations to inform evidence-based investment. Future applications of